Refer to Patient satisfaction Problem 6.15. Consider only the first two predictors (Patient's age, X1. an seventy of illness, X1).
a. Fit a two-region regression tree. What is the first split point, and on which predictor is it based? What is SSE for the resulting two - region tree?
b. Find the second split point given the two-region tree in part (a). Is it is based on X1 or X2 ? What is SSE for the resulting three-region tree?
c. Find the third split point given the three-region tree in part (b). Is it based on X1 or X2. What is SSE for the resulting four-region tree?
d. Find the fourth split point given the four-region tree in part (c). Is it based on X 1 or X2? What is SSE for the resulting five-region tree?
e. Prepare a three-dimensional surface plot of the five-region tree obtained in part (d). What does this tree suggest about the relative importance of the two predictors?
f. Prepare a residual plot of ei versus for the five-region tree in part (d). State your findings.
Problem:
The effects of five dose levels are to be studied in a completely randomized design, and 20 experimental units are available. Each dose level is to be assigned to four experimental units selected at random. Use a table of random digits or a random number generator to make an appropriate randomization of assignments.